Abstract

Joint work with Mauricio Suárez.

One of Mary Hesse’s major contributions to the philosophy of science is her work on analogical reasoning. Since the publication of Models and Analogies in Science (1966), her schematization of the problem has become the disciplinary standard. In this paper we want to bring attention to an element of the schematization which has been left remarkably underexplored, namely the negative analogies. We will argue that there is a host of functions for negative analogies in reasoning, and that, importantly, they have a ‘positive’ inferential potential. We map out three well-known relation-establishing functions of negative analogies – ‘crucial’, irrelevant, and required negative analogies (Bartha 2010, Pero & Suárez 2016, Boesch 2021) – as well as a potentially problematic kind of negative analogy which we term ‘scope-fixing negative analogies’ (from Bailer-Jones 2002). We end by drawing on a case study from clinical practice – the use of the Neer Classification schema for the classification of shoulder fractures. Exploring the ways clinicians reason with this schema, we suggest there are two ways in which negative analogies can serve an inferential function in reasoning, and hence, can be seen as having a positive epistemic potential.